the impartial spectator adam smiths moral philosophy mar 2007

THE IMPARTIAL SPECTATOR
This page intentionally left blank
The Impartial
Spectator
Adam Smith’s Moral Philosophy
D. D. RAPHAEL
CLARENDON PRESS
•
OXFORD
1
Great Clarendon Street, Oxford  
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide in
Oxford New York
Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
Oxford is a registered trade mark of Oxford University Press
in the UK and in certain other countries
Published in the United States
by Oxford University Press Inc., New York
 D. D. Raphael 2007
The moral rights of the author have been asserted
Database right Oxford University Press (maker)
First published 2007

All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
without the prior permission in writing of Oxford University Press,
or as expressly permitted by law, or under terms agreed with the appropriate
reprographics rights organization. Enquiries concerning reproduction
outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the address above
You must not circulate this book in any other binding or cover
and you must impose the same condition on any acquirer
British Library Cataloguing in Publication Data
Data available
Library of Congress Cataloging in Publication Data
Raphael, D. D. (David Daiches), 1916–
The impartial spectator : Adam Smith’s moral philosophy / D.D. Raphael.
p. cm.
Includes bibliographical references and index.
ISBN–13: 978–0–19–921333–7 (alk. paper)
ISBN–10: 0–19–921333–X (alk. paper)
1. Smith, Adam, 1723–1790. 2. Ethics, Modern—18th century.
3. Smith, Adam, 1723–1790. Theory of moral sentiments. I. Title.
B1545.Z7R37 2007
170—dc22 2006036960
Typeset by Laserwords Private Limited, Chennai, India
Printed in Great Britain
on acid-free paper by
Biddles Ltd., King’s Lynn, Norfolk
ISBN978–0–19–921333–7
13579108642
Acknowledgement
Chapters 4–6 are a revised and extended version of the major part of

the Dawes Hicks Lecture on Philosophy given at the British Academy
in 1972, and reproduced here by kind permission from Proceedings
of the British Academy, 58 (1972) ( The British Academy 1973).
This page intentionally left blank
Contents
1. Two Versions 1
2. Sympathy and Imagination 12
3. Motive and Consequences 21
4. Spectator Theory 27
5. The Impartial Spectator 32
6. Comparisons and Comment 43
7. Moral Rules 53
8. Virtue 65
9. The Cardinal Virtues 73
10. Virtue and Beauty 81
11. Ethics and Theology 94
12. Jurisprudence 105
13. Ethics and Economics 115
14. Smith’s Enduring Contribution 127
Bibliography 136
Index 141
This page intentionally left blank
1
Two Versions
Adam Smith is known to the world as the author of The Wealth of
Nations, a pioneering classic in the field of economics. That work
was first published in 1776, when Smith was almost 53 years old.
He wrote the first version of his other book, The Theory of Moral
Sentiments, much earlier: it was published in 1759, when he was a
young professor of 36. A drastically revised and expanded version, the

sixth edition, appeared a few months before Smith’s death in 1790 at
the age of 67. The Moral Sentiments,unlikeThe Wealth of Nations,is
not one of the great classical texts in its field, moral philosophy, but it
has a prominent place among texts of the second rank. Smith himself
is said to have thought it superior to The Wealth of Nations.Despite
some long-winded sentences, the language is hardly ever obscure and
the argument is easy to follow. Yet it has often been misunderstood
and on that account it calls for an interpretation based on knowledge
of what Smith wrote in his youth and in his relative old age.
One source of misunderstanding is that many of the commentators
have been economists who have looked at the Moral Sentiments
simply in order to find some relevance for The Wealth of Nations.
This gave rise to the so-called Adam Smith problem, a supposed
inconsistency between the psychological assumptions of the two
books.
Another source of error has been a failure to note whether a
particular passage was written for the first or for the sixth edition.
Until the publication of the Glasgow Edition of Smith’s works, most
readers of the Moral Sentiments used a copy that reproduces the
text of the sixth edition with no indication that the original version
2 The Impartial Spectator
differed. And even after the Glasgow Edition became available, some
otherwise well-equipped scholars, arguing a case for the views of
Adam Smith, have quoted passages without looking to see whether
they were written for 1759 or 1790.
An example of unfortunate failure to check whether a passage was
written for 1759 or 1790 affects a paper by Professor John Dunn
when he was discussing the ‘practical atheism’ of Smith in his later
years.¹ Dunn contrasts that attitude with, as he thinks, the views
of the youthful Smith who wrote in the Moral Sentiments that ‘the

very suspicion of a fatherless world, must be the most melancholy
of all reflections’. That statement was in fact written by a no longer
youthful Smith for the sixth edition of 1790.
A comparable error was made by Professor Jacob Viner in an
important article on laissez-faire in Smith’s economics.² He contras-
ted the mature realism of The Wealth of Nations with the youthful
idealism of the Moral Sentiments, and quoted five passages from the
ethical work as evidence for his view of it. The first of his quotations
was in fact written for the far from youthful sixth edition.
A third example is a lapse in a perceptive interpretation of the
Moral Sentiments by Professor Charles L. Griswold, bringing out the
influence of drama in Smith’s book.³ He claims that, when Smith
writesofthespectator’smoraljudgement,heenvisagesthespectator
of a dramatic performance seeing the agent as an ‘actor’ on the stage.
The evidence that Griswold adduces is one instance of the word
¹ John Dunn, ‘From applied theology to social analysis: the break between John
Locke and the Scottish Enlightenment’, in Istvan Hont and Michael Ignatieff (eds.),
Wealth and Virtue (Cambridge: Cambridge University Press, 1983), 119, 128. Dunn is
in fact aware (p. 120 n.) that the sixth edition of the Moral Sentiments differs significantly
from the earlier versions, but by an unfortunate lapse he attributes to 1759 his quotation
about a fatherless world (p. 128).
² Jacob Viner, ‘Adam Smith and Laissez Faire’, Journal of Political Economy,35
(1927), 198–232; repr. in Viner, The Long View and the Short (Glencoe, Ill.,
1958).
³ Charles L. Griswold, Jr., Adam Smith and the Virtues of Enlightenment (Cambridge:
Cambridge University Press, 1999), 65–6, 87n. In discussing the theatrical character
of the Moral Sentiments Griswold is much influenced by David Marshall, TheFigureof
Theater (New York: Columbia University Press, 1986). Marshall’s chapter on Smith is
naturally concerned only with this feature of Smith’s book.
Two Versions 3

‘actor’ in place of ‘agent’ in the Moral Sentiments and one instance
in the Lectures on Rhetoric.TheMoral Sentiments instance occurs in
part VI of the book, which was added to the sixth edition of 1790; so
Griswold’s evidence cannot apply to Smith’s general conception of
the spectator. There was in fact an instance of the word ‘actor’ in the
first edition which was replaced by ‘agent’ in subsequent editions,
showing that there was clearly no association with actors on the
stage. This flaw in Griswold, however, does not lessen the value of
his interpretation as a whole.
Having criticized those three scholars, I should add that sever-
al other recent commentators on Adam Smith, Professors Knud
Haakonssen, Gloria Vivenza, Vivienne Brown, Stephen Darwall,
Emma Rothschild, and Samuel Fleischacker, do take account of
differences between the first and the sixth editions of the Moral
Sentiments.⁴ Haakonssen and Fleischacker take account also of a
surviving fragment of a lecture giving a still earlier version of Smith’s
treatment of justice.⁵ Fleischacker’s book is mainly focused upon
The Wealth of Nations butincludessomesubtleanalysisoftheMoral
Sentiments—and indeed of the Lectures on Jurisprudence known to
us from student reports.
An earlier commentator, Professor T. D. Campbell, was well aware
of differences in the various editions of the Moral Sentiments and
takes note of them in his book Adam Smith’s Science of Morals.⁶ Like
⁴ Knud Haakonssen, The Science of a Legislator (Cambridge: Cambridge University
Press, 1981), 150–1, 217; Gloria Vivenza, Adam Smith e la cultura classica (Pisa: Il
pensiero economico moderno, 1984), 63–5; English version, revised and enlarged, Adam
Smith and the Classics (Oxford: Oxford University Press, 2001), 54–5, 57; Vivienne
Brown, Adam Smith’s Discourse (London and New York: Routledge, 1994), 134–40;
Stephen Darwall, ‘Sympathetic Liberalism: Recent Work on Adam Smith’, Philosophy
and Public Affairs, 28 (1999), 153–4; Emma Rothschild, Economic Sentiments: Adam

Smith, Condorcet, and the Enlightenment (Cambridge, Mass., and London: Harvard
University Press, 2001), ch. 5; Samuel Fleischacker, On Adam Smith’s Wealth of Nations
(Princeton: Princeton University Press, 2004), 48, 83, 112–14, 148–9.
⁵ See appendix II of the Glasgow Edition of Adam Smith, TheTheoryofMoral
Sentiments (Oxford: Clarendon Press, 1976; corrected reprint, 1991), 389, 397. In my
subsequent notes the work is cited as TMS.
⁶ T. D. Campbell, Adam Smith’s Science of Morals (London: George Allen &
Unwin, 1971).
4 The Impartial Spectator
Fleischacker, he also considers relevant material from the lectures on
jurisprudence, which he read in the actual student manuscript before
it was reproduced in print. Tom Campbell is a former colleague of
mine, and his book on Adam Smith is a revised version of a Ph.D.
thesis that he wrote under my supervision, though I may say that he
needed very little supervising, so that the book owes virtually nothing
to me. I refrained from rereading it before writing this book lest it
might affect what I had to say; I wanted to stick to my own thoughts,
which have arisen from frequent reading of Smith’s work as an editor.
Having reread Campbell’s book now in the final stage of preparing
my own book, I have found that there is in fact very little difference
between Campbell’s interpretation and mine. There is, however,
enough difference between the character of the two books to justify
their separate existence. Campbell’s book emphasizes Smith’s aim to
produce a work of science and discusses the moral philosophy as a
part of that aim. My account goes into more detail on the particular
content of the Moral Sentiments and suggests that the concept of the
impartial spectator is especially concerned with moral judgements
about one’s own actions.
In contrast to my substantial agreement with Campbell, I have
some criticisms to make of the views of Viner, Brown, and Fleis-

chacker, and I shall return to them in Chapter 13. The references
by Haakonssen, Vivenza, Darwall, and Rothschild to different edi-
tions of the Moral Sentiments do not call for comment. Vivenza’s
book is notable for its illuminating account of the influence of
Greek and Roman thought, especially Stoicism, on Adam Smith.
That is an important topic but one on which I have no particular
competence and which I have therefore left alone. Haakonssen is
chiefly concerned with jurisprudence. I am told that Darwall has
a more substantial discussion of Smith’s ethics in a book, Second
Person Standpoint, due to be published shortly. Rothschild’s discus-
sion of Adam Smith is mainly about The Wealth of Nations and
includes a chapter on the invisible hand, in which she is well aware
of the differences between the first and sixth editions of the Moral
Two Versions 5
Sentiments. I am not altogether persuaded by her ingenious argument
that Smith’s use of the invisible hand is ironic, but it does not affect
the essentials of his ethical theory.
Let me recall briefly the facts of publication of the Moral Senti-
ments. The first edition appeared in 1759. A second edition, with
revisions of some substance, was published in 1761, and was followed
by third, fourth, and fifth editions in the years 1767, 1774, and 1781.
Revisions in those three editions were light, though not negligible in
the case of the third edition. All these changes, however, including
those of the second edition, are minor matters when compared with
the sixth edition, which was submitted to the printer at the end
of 1789 and published in May 1790. The sixth edition includes a
whole new part, on the character of virtue, and some drastic revision
elsewhere. My co-editor of the work, the late Professor A. L. Macfie,
was apt to say in consequence that the sixth edition is a different
book. That is an exaggeration, but the sixth edition is certainly a

much altered book.
The primary purpose of the work is to expound a theory of
ethics. In saying this I do not rely on the title, The Theory of Moral
Sentiments, though that does tend to confirm what I have just said.
The title is not meant to be a name for Smith’s own theory: rather,
it is a name for the subject matter, as we may see from the surviving
manuscript fragment (mentioned above) of Smith’s lecture on justice
at Glasgow University. He writes there of the rules that constitute
‘what is called Natural Jurisprudence, or the Theory of the general
principles of Law. they make a very important part of the Theory
of moral Sentiments.’ That, of course, implies that Smith’s own
contribution is an essay in theory. However, the chief evidence of
Smith’s primary purpose is the content of the book: by far the largest
component of it is philosophical analysis.
I stress the point because a number of commentators have laboured
to derive from the book Smith’s personal stance on moral questions;
and at least one well-informed commentator, Professor Griswold,
has emphasized Smith’s ‘protreptric’ purpose, that is to say, his desire
6 The Impartial Spectator
to promote the practice of virtue.⁷ There is nothing wrong with such
an approach to Smith’s work. Philosophers who write on ethics do
often have a particular personal stance on some moral questions, and
when one of those philosophers is a famous world-figure for other
reasons, it is both natural and legitimate to seek to elicit his character
andpersonalityfromhiswritingsaswellashisactions.Andwhile
some philosophers, notably Smith’s friend David Hume,⁸ think that
the philosophical explanation of ethics is muddied by mixing it with
the promotion of morality, it is true enough that Smith goes in for
this—though far more in the new part VI of the sixth edition than in
the original work: you will find little of it if you read the first edition.

The primary object of the book in all editions is to contribute to
ethical theory.
When Smith wrote that natural jurisprudence is part of the theory
of moral sentiments, he could just as well have said ‘part of moral
philosophy’. That is what he meant, taking ‘moral philosophy’ in
the wide sense that it had for the Scottish universities of his time.
Why does he call it the theory of moral sentiments? To answer that
question one needs to recall the recent previous history of the subject.
Francis Hutcheson and David Hume were the two most prom-
inent Scottish contributors to moral philosophy before Smith. They
had criticized the view of rationalist philosophers, such as Samuel
Clarke and William Wollaston, that the judgement and the motive of
moral action are functions of reason, an understanding of necessary
truth analogous to mathematical thinking. Hutcheson and Hume,
in contrast, took the view that moral judgement is affective, rests on
feeling, and that the motive for acting upon that judgement must
likewise be affective, since reason alone does not have the power to
stir bodily behaviour. Hume was a particularly trenchant critic: he
began his discussion of morals in book III of his Treatise of Human
⁷ Griswold, Adam Smith, ch. 1, §2, and epilogue, p. 366.
⁸ Letter of 17 September 1739 to Hutcheson; Letters of David Hume,ed.J.Y.T.
Greig (Oxford: Clarendon Press, 1932), i. 32–3. Hume is referring to the manuscript of
book III of ATreatiseofHumanNature. Some scholars think that he did not maintain
this opinion in his later work.
Two Versions 7
Nature with a battery of arguments to show that ‘moral distinctions’
are ‘not deriv’d from reason’ and concluded that they are ‘deriv’d
from a moral sense’. He borrowed the term ‘moral sense’ from
Hutcheson, but used it only in the title of the relevant section; in the
body of his discussion he wrote instead of ‘feeling’ or, much more

often, of ‘sentiment’.
I think Adam Smith took it for granted that Hume had demon-
strated beyond challenge his conclusion that moral distinctions arise
from feeling. Smith therefore proceeded on the assumption that any
further contribution to moral philosophy must make ‘sentiment’, in
the sense of feeling, the basic element of its account.
ThescopeofSmith’scontributionisrelativelynarrow.Itsmain
concern is the nature of moral judgement, as is recognized in a lengthy
subtitle that first appears in the fourth edition. The earlier editions
had borne only the main title, ‘The Theory of Moral Sentiments’,
but the fourth edition is more explicit: ‘The Theory of Moral
Sentiments, or An Essay towards an Analysis of the Principles by
which Men naturally judge concerning the Conduct and Character,
first of their Neighbours, and afterwards of themselves’. The analysis
is of a matter of fact, the principles (general rules) that human
beings do in fact follow when they pass judgement on conduct and
character.
It is an explanation in terms of psychology and sociology. Adam
Smith does not, like his Scottish predecessors, describe his project
as an inquiry into the ‘original’ (Hutcheson) or ‘origin’ (Hume) of
ethical ideas and judgements, but all three philosophers are doing
the same thing, seeking a genetic explanation. Hutcheson’s inquiry is
conducted almost entirely in terms of psychology. Hume follows suit
in the main, but brings in sociology at times—for example, when he
draws an important distinction between natural and artificial virtues.
With Adam Smith sociology looms larger. This is not to say that
psychology recedes into the background: the psychological element
in Smith’s explanation of ethics is vivid, often strikingly original,
and usually persuasive. It is, however, all the more illuminating for
being allied with acute sociological observation.

8 The Impartial Spectator
You might say that this must limit the character of Smith’s
explanation of ethics: his evidence is drawn from his own society and
lacks the universality that is sought by philosophers who reflect on
general human experience rather than the experience of a particular
section of mankind at a particular period of time. This suggestion
cannotbedismissed,butitisoverdrawn.
Smith certainly thought he was reflecting on general human
experience, and his evidence was not limited to his own society.
He had a fairly wide knowledge of history, including the history
of ancient Greece and Rome, and he took a keen interest in such
anthropological reports as were available, notably reports about
the indigenous inhabitants of North America. He knew that the
behaviour and the ethics of the American Indians differed markedly
in some respects from what was found in Europe, and he knew,
too, that the ethics of ancient Graeco-Roman civilization differed,
in other respects, from the ethics of Christianity. He was also aware
of minor differences in the mores of at least some European societies
(France and Italy as compared with Britain and with each other); his
description of these differences is given in the first edition of 1759,
before he had set foot on the continent of Europe so as to see for
himself what went on in France.
Despite all these sociological observations on variations in beha-
viour and moral outlook, Smith still thought he could appeal to an
agreed consensus among his reflective readers on the relative merits
of differing codes of conduct. He may have been too sanguine in
supposing that this consensus had a universal truth. Still, even if the
scope of his explanation may be limited, it remains enlightening for
us today, since our ethics are not radically different from the ethics
of Adam Smith’s time and place.

To be sure, there have been some changes of attitude. For example,
there is less deference now to ‘the rich and the great’ than there was
in Smith’s day, and a weaker trace of the ancient ethic of honour
and shame that Smith finds among the ‘gallant and generous part
of mankind’ (TMS . iii. 1. 15). But such changes do not cast
doubt upon Smith’s explanation of moral judgement, because the
Two Versions 9
sociological facts on which he relies for his explanation are of a more
general nature.
He tells us that moral approval is related to the sympathetic
and antipathetic feelings of spectators. If the use of those feelings
changes, in respect of social status or anything else, so does the
use of moral judgement: a reduction or cessation of deference to
the rich and the great implies that one no longer feels a special
moral obligation to meet their wishes. But the general thesis, that
moral judgements depend on the feelings of spectators, remains as
before. It seems impossible to imagine a set of human beings whose
moral judgements are not linked to general social attitudes. Even
in the fantastic fictional society of Samuel Butler’s Erewhon,where
moral condemnation is applied to illness while criminal behaviour is
greeted with commiseration, both classes of judgement depend on
shared social attitudes. In Hobbes’s fictional state of nature, human
beings are represented as influenced solely by egoistic aims; but the
consequence of that, Hobbes tells us, is a complete absence of right
and wrong.
Towards the beginning of the final part of the Moral Sentiments
Smith writes:
In treating of the principles of morals there are two questions to be
considered. First, wherein does virtue consist? Or what is the tone of
temper, and tenour of conduct, which constitutes the excellent and praise-

worthy character, the character which is the natural object of esteem,
honour, and approbation? And, secondly, by what power or faculty in
the mind is it, that this character, whatever it be, is recommended to us?
Or in other words, how and by what means does it come to pass, that
the mind prefers one tenour of conduct to another, denominates the one
right and the other wrong; considers the one as the object of approbation,
honour, and reward, and the other of blame, censure, and punishment?
(TMS .i.2)
One would expect such a programmatic statement to come at the
beginning of the book, not near the end. It stands at the beginning of
a history of moral philosophy, specifically a history of theories about
the two questions that Smith says constitute the subject. There is
10 The Impartial Spectator
plenty of evidence that both of Smith’s books were developed from
his lectures. It is also plain that his usual method of inquiry was to
begin with the history of his subject and to reach his own views in
the light of his survey of history. I have therefore conjectured⁹ that
the passage quoted above was originally the beginning of his lectures
on moral philosophy and that the lectures then continued with the
critical survey of history before turning to Smith’s own theory.
Smith’s own theory, as given in the first five editions, is for the
most part a theory of moral judgement—that is to say, it is an answer
to the second question set out in the initial description of the subject
of philosophical ethics. I do not say that it contains nothing about
the first question, the character of virtue: there is a relatively short
discussion of a distinction between the ‘amiable’ and the ‘respectable’
virtues, summarized as sensibility and self-command, and a longer
discussion of the contrast between justice and beneficence. But there
is no thoroughgoing inquiry of what constitutes the character of
virtue, as required by the first of the two questions, even though the

historical survey at the end of the book deals with both questions
in turn and, as it happens, gives more space to the first topic,
the character of virtue, than to the second, the nature of moral
judgement.
The fact is that Smith did not reach a distinctive view on the first
topic. He has a distinctive view of the content of virtue, that is to
say, a view of what are the cardinal virtues; but he does not give us
an explanation of what is meant by the concept of moral virtue, how
it arises, how it differentiates moral excellence from other forms of
human excellence. The main subject matter of the first version of the
book is well described in the long title added for the fourth and later
editions: it is a detailed explanation of moral judgement, as passed
first on the actions of other people and then on the actions of oneself.
I think that, when Smith came to revise the work for the sixth edition,
he realized that he had not dealt at all adequately with the first of
the two questions, and for that reason he added the new part VI,
⁹ TMS .i.2,editorialn.1.
Two Versions 11
entitled ‘Of the Character of Virtue’, to remedy the omission. It
is not, in my opinion, an adequate remedy, and it certainly does
not match Smith’s elaborate answer to the second question. It does,
however, bring out the exceptional role of self-command among the
virtues and thereby shows Smith’s theory of virtue to be a distinct
advance on that of his predecessors. Self-command has a place in the
earlier version, as a marker of the ‘respectable’ (as contrasted with
the ‘amiable’) virtues; in the later version it looms larger, being the
determinant of a superior form of any virtue.
Since the second of the two topics, the nature of moral judgement,
is the main subject of both versions of Smith’s book, I shall give
it priority in what follows. There is in fact a clear development

in Smith’s view of this topic, especially in his conception of the
impartial spectator, the most important element of Smith’s ethical
theory. Hence the title of my book.
2
Sympathy and Imagination
The first chapter of The Theory of Moral Sentiments is entitled ‘Of
Sympathy’, and the first chapter of An Inquiry into the Nature and
Causes of the Wealth of Nations is entitled ‘Of the Division of Labour’.
In each instance, I think, the title is meant to indicate the primary
cause of the subject matter of the book: the moral sentiments are
founded on sympathy, and the increase of national wealth is founded
on the division of labour. So far as the Moral Sentiments is concerned,
the name of the primary cause has a wider sense than you might
suppose. Smith notes that in common usage the term ‘sympathy’
tends to be limited to pity, fellow-feeling with distress, and he makes
a point of telling us that he is using the term, as its etymology allows,
to mean the sharing of any kind of feeling.
If, as I have suggested, the title of the first chapter is intended
to pinpoint the primary cause of the book’s theme, ‘imagination’
should be added to ‘sympathy’; and of the two, imagination plays
the larger part. In saying this I do not imply that sympathy is
always accompanied by imagination, that sympathy, as understood
by Smith, cannot get going until one has consciously imagined
oneself into the shoes of another person. Smith’s first examples
of sympathy seem to belie that idea: they describe a spontaneous
repetition of feeling and observed behaviour.
When we see a stroke aimed and just ready to fall upon the leg or arm of
another person, we naturally shrink and draw back our own leg or our own
arm; and when it does fall, we feel it in some measure, and are hurt by it
as well as the sufferer. The mob, when they are gazing at a dancer on the

slack rope, naturally writhe and twist and balance their own bodies, as they
Sympathy and Imagination 13
see him do, and as they feel that they themselves must do if in his situation.
Persons of delicate fibres and a weak constitution of body complain, that in
looking on the sores and ulcers which are exposed by beggars in the streets,
they are apt to feel an itching or uneasy sensation in the correspondent part
of their own bodies. (TMS .i.1.3)
Smith himself does not distinguish such examples from explicitly
imagining oneself in the place of another person. He thinks that
imagination is involved in almost all instances, and he gives the
examples of the above quotation to illustrate the fact that our
awareness of the feelings of other people can only come from
imagining ourselves in their shoes and seeing what we would then
feel. Smith does, however, go on to mention other examples of
sympathy that are entirely spontaneous and are not accompanied by
theexerciseofimagination.
Upon some occasions sympathy may seem to arise merely from the view
of a certain emotion in another person. The passions may seem to be
transfused from one man to another, instantaneously Grief and joy, for
example, strongly expressed in the look and gestures of any one, at once
affect the spectator with some degree of a like painful or agreeable emotion.
A smiling face is, to every body that sees it, a cheerful object; as a sorrowful
countenance, on the other hand, is a melancholy one. (TMS .i.1.6)
Whether we think that sympathy without imagination is confined
to the second set of examples or occurs in the first set also, it is
a fairly unusual phenomenon. On most occasions imagination is a
prerequisite for sympathy.
An explicit exercise of the imagination is certainly part of Smith’s
account of moral judgement. In that context imagining oneself in
someone else’s place is more pervasive than the actual experience of

sympathy.
Let us consider first Smith’s account of the judgement that an
action is proper or improper. He writes also of the judgement that an
affection, a feeling, is proper or improper, and here he is not confining
himself to the feeling, the motive, of an agent: he would include the
feeling of the person or persons affected by an action, and also the
14 The Impartial Spectator
feeling of a person affected by an event, such as the death of a relative
or friend. It is, however, convenient to begin with the judgement
that an action is proper. In principle though not in precise detail,
Smith is talking about the simple judgement that an action is right
or wrong. Smith calls it a judgement of propriety or impropriety, an
assertion that an action is appropriate or inappropriate, suitable or
unsuitable, to the cause that has prompted the agent to do it. The
primary form of such judgement, according to Smith, is made by a
spectator on an action done or contemplated by another person. The
spectator’s judgement arises from imagining himself in the agent’s
place and comparing the motivating feeling of the agent with the
feeling that he himself would have in the imagined situation. If his
own imagined feeling is the same as the actual feeling of the agent, he
is ‘sympathizing’ with the agent, and his awareness of the sympathy
(fellow-feeling) is given expression in approval, declaring that the
action is appropriate (right). If, on the other hand, his own imagined
feeling differs from that of the agent, he lacks sympathy with the
agent, and his awareness of this is given expression in disapproval,
declaring that the action is inappropriate (wrong).¹
Smith does not confine disapproval to a positive feeling of anti-
pathy (a word that he uses very rarely), but seems to think that
any degree of difference in feeling will give rise to disapproval of
the action concerned. The possibilities for disapproval are therefore

manifold in kind, though not necessarily more frequent than the
occurrence of approval. However, the exercise of imagination is
required for all the possibilities, while the experience of sympathy is
¹ One of the Press’s advisers says some scholars claim that, ‘for Smith, the very act
of imaginative identification is itself an act of ‘‘sympathizing’’ with the agent’, and he
asks for textual evidence to the contrary. In his first chapter Smith defines sympathy as
fellow-feeling and says that ‘changing places in fancy with the sufferer’ is ‘the source of our
fellow-feeling’ (emphasis added). Changing places in fancy is an act of the imagination;
if it is the source of fellow-feeling, it cannot be itself the fellow-feeling. The adviser gives
only one name, saying that Charles Griswold, ‘at moments, seems to be one such’, citing
Griswold’s reference to Smith’s example of sympathy with the dead. I think Griswold
takes my view. He writes (Adam Smith, 90) of ‘This ‘‘illusion of the imagination’’ thanks
to which we sympathize with the deceased’ (emphasis added): the sympathy is an effect
of the act of imagination.
Sympathy and Imagination 15
confined to only one of them. Hence I say that imagination is more
pervasive than sympathy in the forming of moral judgements.
There are further reasons for saying that Smith thinks of imagin-
ation as more pervasive than feelings of sympathy. He notes that
the moral reflection of a spectator often does not depend on any
actual perception of a correspondence with, or difference from, the
feeling of the agent. ‘We sometimes feel for another, a passion of
which he himself seems to be altogether incapable; because, when
we put ourselves in his case, that passion arises in our breast from
the imagination, though it does not in his from the reality’ (TMS
. i. 1. 10). We feel compassion for the deranged (or even for the
dead), because, by an ‘illusion of the imagination’ (TMS .i.1.13),
we attribute to them feelings of distress which they do not have but
which we suppose that we, being rational instead of deranged (or
being alive and conscious instead of dead), would have if we were in

their situation. In these examples, as Smith portrays them, there is
compassion for the deranged or the dead, but it is not a sympathetic
compassion, because we know that the objects of our compassion are
not in fact feeling distress; we are illusorily imagining a distress that
we, retaining our present faculties, would feel in their situation.
Later, when he comes to deal with moral judgements on our
own conduct, Smith gives the imagination an elaborate double role:
we have to imagine what spectators would feel if they imagined
themselves in our situation; and, while sympathy, or the lack of it,
comes into the picture in characterizing the feeling of the spectators,
that feeling is an imagined feeling; and indeed, in the end, spectators
in the real world are replaced by an imagined impartial spectator
conjured up ‘in the breast’.
I come back to the starting point. A spectator observes or hears of
an action done or contemplated. He knows its ‘cause’; that is to say,
he knows what has prompted the agent to act or think of acting. Let
us suppose that the agent has come upon a child struggling to swim
to the bank of a river; he dives in to help. The ‘spectator’ imagines
himself in the agent’s place and notes that he would be prompted
to act likewise. In other words, he finds that he ‘sympathizes’ with
16 The Impartial Spectator
the agent’s feelings and consequent action. He gives expression to
his sympathy by approving of the action as right or proper, an
appropriate response to the situation.
Now suppose instead that the agent cannot himself swim. There
is no point in his diving into the water; that would not help the
child and would simply add a second person in danger of drowning.
He cannot see a lifebelt or a rope or another passer-by, and so he
takes off his shirt and uses that as a sort of rope. Let us hope that it
works; anyway he thinks it is the best he can do. The ‘spectator’, who

hears about the episode, imagines himself in the agent’s shoes—and
sharing the agent’s inability to swim. He finds, reluctantly but
inescapably, that here again he would be prompted to act in much
the same way as the agent, and so he approves of the action as
appropriate to the situation.
Let us now suppose that the agent meets someone who, having
heard a garbled account of the incident, accuses him of cowardice.
The agent, aggrieved at the taunt, punches the scoffer on the nose.
When the ‘spectator’ learns of this and imagines himself in the
agent’s shoes, he finds that he too would feel aggrieved but would
not be disposed to respond with a punch. Since he does not fully
sympathize, he disapproves of the punch and says it was wrong,
inappropriate to the situation.
The example, with its two components, illustrates the base of
Smith’s theory. Moral judgement begins with the reaction of spec-
tators to the actions and motives of other people. The ‘spectators’ in
question are normal fellow-members of society. Smith assumes that
nearly all of them will react in much the same way. They include
you and me, and for the most part Smith writes of what ‘we’ feel
and think about the conduct of other people. But in key passages
of his explanation he writes of ‘the spectator’ (occasionally ‘spectat-
ors’), because the relevant feelings and thoughts that ‘we’ experience
have come to us in our capacity as spectators. He sometimes writes
of ‘mankind’ or ‘every body’, but he knows that unanimity can-
not always be guaranteed and so he sometimes introduces a slight
qualification, as in ‘every impartial spectator’ or ‘every indifferent